EP0417782A2 - Adaptive process control system, especially for control of temperature of flowing fluids - Google Patents
Adaptive process control system, especially for control of temperature of flowing fluids Download PDFInfo
- Publication number
- EP0417782A2 EP0417782A2 EP90117651A EP90117651A EP0417782A2 EP 0417782 A2 EP0417782 A2 EP 0417782A2 EP 90117651 A EP90117651 A EP 90117651A EP 90117651 A EP90117651 A EP 90117651A EP 0417782 A2 EP0417782 A2 EP 0417782A2
- Authority
- EP
- European Patent Office
- Prior art keywords
- variable
- control
- control point
- response
- changes
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 239000012530 fluid Substances 0.000 title claims description 42
- 230000003044 adaptive effect Effects 0.000 title abstract description 28
- 238000004886 process control Methods 0.000 title description 11
- 238000000034 method Methods 0.000 claims abstract description 112
- 230000008569 process Effects 0.000 claims abstract description 94
- 230000004044 response Effects 0.000 claims abstract description 67
- 230000008859 change Effects 0.000 claims description 32
- 230000001276 controlling effect Effects 0.000 claims description 24
- 238000013459 approach Methods 0.000 claims description 7
- 230000001105 regulatory effect Effects 0.000 claims description 6
- 238000011144 upstream manufacturing Methods 0.000 claims description 3
- 239000007788 liquid Substances 0.000 abstract description 14
- 230000006978 adaptation Effects 0.000 description 22
- 230000006870 function Effects 0.000 description 18
- 230000033228 biological regulation Effects 0.000 description 10
- 239000000126 substance Substances 0.000 description 9
- 238000005070 sampling Methods 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 5
- 238000012937 correction Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 230000003094 perturbing effect Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000001816 cooling Methods 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 230000007423 decrease Effects 0.000 description 2
- 238000012546 transfer Methods 0.000 description 2
- 230000001052 transient effect Effects 0.000 description 2
- 208000034809 Product contamination Diseases 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 230000003466 anti-cipated effect Effects 0.000 description 1
- 238000009529 body temperature measurement Methods 0.000 description 1
- 238000001311 chemical methods and process Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 239000003153 chemical reaction reagent Substances 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 239000011248 coating agent Substances 0.000 description 1
- 238000000576 coating method Methods 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000010438 heat treatment Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000009428 plumbing Methods 0.000 description 1
- 238000010926 purge Methods 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 230000006641 stabilisation Effects 0.000 description 1
- 238000011105 stabilization Methods 0.000 description 1
- 238000009966 trimming Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D23/00—Control of temperature
- G05D23/19—Control of temperature characterised by the use of electric means
- G05D23/20—Control of temperature characterised by the use of electric means with sensing elements having variation of electric or magnetic properties with change of temperature
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D23/00—Control of temperature
- G05D23/19—Control of temperature characterised by the use of electric means
- G05D23/1927—Control of temperature characterised by the use of electric means using a plurality of sensors
- G05D23/193—Control of temperature characterised by the use of electric means using a plurality of sensors sensing the temperaure in different places in thermal relationship with one or more spaces
- G05D23/1931—Control of temperature characterised by the use of electric means using a plurality of sensors sensing the temperaure in different places in thermal relationship with one or more spaces to control the temperature of one space
Definitions
- the present invention relates to process control systems, and more particularly to a control system (method and apparatus) which is adaptive on-line to changing process conditions.
- the invention is especially suitable for use in a process where the temperature of a flowing liquid requires regulation with a fast dynamic response to handle temperature transients while adapting to changing process conditions such as the specific heat of the flowing liquid, or other conditions affecting the temperature of the liquid.
- the process may be part of the coating of photographic paper or film which requires precise temperature regulation.
- the invention is applicable generally wherever a variable of a process is to be controlled, whether that variable be temperature or some other physical or chemical process variable; temperature control of flowing fluids as described herein being an illustrative and presently preferred embodiment of the invention.
- Heat exchangers as are available are inadequate in providing the dynamic response for fast and precise regulation of the temperature of the fluid, for example, within a few seconds.
- heat exchangers generally are large in size, do not heat uniformly and must be maintained by cleaning and purging in order to keep them operative and avoid product contamination.
- Conventional control systems using heat exchangers for temperature control of flowing fluids have been inadequate for providing precise regulation while at the same time adapting (preferably in real time and on-line) to changing process conditions. Such conditions may, for example, be the specific heat of different batches of the fluids used in the process, such as the different batches of a chemical or different chemicals used in the manufacture of photographic products.
- Systems having heating means with faster dynamic response than heat exchangers and conventional control systems have been inadequate to provide the precise regulation which is needed by adapting to changing process conditions or compensating on-line and continuously for other changing factors which require a change in the dynamic response of the control system.
- Control systems sometimes have both feedforward and feedback control loops; feedforward control being responsive to cancel an incoming disturbance or change while feedback control being responsive to the result of the action which has been taken.
- Feedback and feedforward controllers have been combined, the feedback controller compensating for any error remaining after the feedforward action. See for example the following patents discussing feedback and feedforward controllers: Harrison, et al., U.S. 3,976,179, August 24, 1976; Koschmann, U.S. 4,290,986, September 22, 1981; Dibiano, U.S. 4,367,354, January 4, 1983; Peterman, et al., U.S. 4,543,245, September 24, 1985 and Dziubakowski, 4,776,301, October 11, 1988.
- adaptive control is implemented by perturbing the system ( See Paz, U.S. 3,567,395 of March 2, 1971). Then the system response is detected and the system is adjusted or tuned.
- This approach is inadequate when the effect of these perturbations produces a variation beyond the limits of regulation required. For example, in a system requiring precise regulation of temperature of flowing fluids, particularly chemicals in a photographic film manufacturing process, the perturbation could change the temperature beyond the limits required by the process.
- the approach is difficult to implement since both the amplitude and frequency content of the perturbing input must affect all response modes of the controlled process which are of interest. Otherwise, the control algorithm adaptation can diverge rather than converges. In other words, the system never reaches a state where the controlled variable is within the desired range, but rather is forced out of control; certainly an undesirable result.
- the microwave generator is under feedforward and feedback control so as to precisely regulate the temperature.
- Another feature of the invention is to provide control of a variable using feedback and feedforward loops with adaptive control of the feedforward control function from the feedback loop.
- the feedforward controller (if conditions were ideal, which is not the case) would substantially eliminate any deviation of the controlled variable from the desired value.
- the feedback controller acting as a backup, then reduces the remaining deviation, and the output from the feedback controller is a measure of the feedforward controller's inability to reduce the remaining deviation (called tuning error) when the system reaches equilibrium.
- tuning error the remaining deviation
- the feedback output information has, in accordance with the invention been found sufficient to adapt or tune the feedforward controller to match changing process conditions, without having to perturb the controlled variable in order to derive the necessary adaptation or tuning signals.
- An additional advantage of feedback control of adaptation, in accordance with the invention, is that its implementation is far less complicated than with approaches requiring perturbation of the controlled variable. Since the controlled variable does not have to be perturbed, the system is tuned without driving the variable outside of the limits in which it must be maintained to meet process requirements.
- the process variable may be temperature of a flowing fluid and a microwave generator which has an extremely fast response is responsive to the output of one or both of the controllers.
- the invention which may be embodied in a control system for temperature control by the application of microwave energy from a microwave generator which provides heat which changes the controlled variable.
- the system also includes feedforward and feedback controllers respectively responsive to the variable before and after a control point where the heat is applied from the microwave generator.
- the adaptive regulation of a process variable in accordance with the invention is carried out in a method having the following steps each of which may be performed with a digital computer.
- the steps are measuring the variable before and after a control point in the process; controlling the variable at the control point in response to changes in the variable as measured before the control point in accordance with a predetermined control function which relates the variable to the conditions of the process; and adapting the control function in response to changes in the variable as measured after the control point and when the changes represent changes in the condition of the process.
- the condition of the process being represented by changes, for example, when the variable is temperature, when the temperature error, as detected by the feedback controller, has such a low magnitude and slow rate of change as occurs when the process approaches stabilization.
- the adaptive process control system has a feedforward controller.
- this controller has proportional and derivative responses.
- only the proportional response is adapted to match changing process conditions.
- the derivative response or both proportional and derivative responses can be adapted.
- the system has a feedback controller having proportional and integral responses.
- the feedforward response may change in one sense, for example, to increase the temperature where the temperature is the controlled variable, while the feedback controller output can change in opposite senses.
- the feedforward conroller has a unipolar response whereas the feedback controller has a bipolar response which can add or subtract from that of the feedforward controller.
- Adaptation is obtained by changing the gain (a constant in the control function) of the feedforward controller in accordance with the feedback output.
- the system operates iteratively, and is implemented with a digital computer having a sampling rate appropriate to accurately detect the fastest rate of change anticipated in the controlled variable.
- a sampling rate of one second may be used.
- the feedforward gain is adjusted to tune the feedforward controller at steady state conditions when the output rate and magnitude of the variable to which the feedback loop is responsive (the temperature after the control point) is well within limits within which regulation is desired. For example, in a system which regulates temperature within a ⁇ 0.5°F, the limits may be 0.25°F in magnitude and 0.025°F per second in rate of change.
- FIG. 1 there is shown apparatus for carrying out a portion of a photographic film manufacturing process.
- the chemical being dispensed is a fluid (a liquid) which must be regulated precisely in temperature.
- the chemical is mixed in kettles (not shown) and passes through a system of plumbing including valves and other hardware which are also not shown.
- the portion of the process shown in FIG. 1 provides the necessary temperature control.
- the chemical flows through a plastic pipe 10 past a temperature control point 12 in a wave guide applicator section 14, which applies microwave energy from a microwave generator 16 to heat the chemical at the control point.
- Conductive couplings 18 and 20 are used to seal the applicator 14 against the escape of microwave energy.
- a plastic pipe 10 carrying the fluid extends through these couplings.
- the applicator 14 is tuned to the frequency of the microwave generator.
- a suitable generator is made by Gerling Laboratories of Modesto, California, USA and produces an output frequency of 2.45 GHz. Other frequencies may be used, for example, when larger waveguides are practical.
- the microwaves are transmitted by a wave guide 22 to the applicator 14.
- the applicator 14 is tuned by a slug tuner 24. Such tuners are shown in U.S. Patent 4,689,459 issued August 25, 1987.
- the microwave generator is controlled by a control signal (for example, a voltage which may vary from 0 to minus 1 volt) which changes the microwave energy supplied to the applicator 14 from 0 to 3 KW.
- the temperature of the incoming fluid is measured ahead of the control point and also after the control point 12 by temperature sensors 26 and 28, which are close, say within one foot, of the control point so that time delay between temperature measurements is minimal and the temperature at the sensor 26 is substantially the same as the temperature of the fluid at the control point.
- the sensors are suitably resistance temperature devices (RTD) which are responsive to the temperature of the liquid in the pipe.
- RTD resistance temperature devices
- the RTD's form one arm of bridge circuits. These bridge circuits are contained in RTD transmitters 30 and 32.
- the transmitters also contain amplifiers which produce analog outputs, for example, currents (e.g. from 4 to 20 mA) proportional to the temperature measured by the RTDs 26 and 28. These analog outputs are indicated as E Tin and E Tout .
- a flow sensor 34 is also disposed to sense the flow rate of the liquid through the pipe 10.
- This flow sensor may be a Micromotion flow sensor.
- Such sensors have a vibrating loop with pickup coils. The loop vibrates at a frequency of vibration proportional to the rate of flow through the loop. Other sensors may be used such as vane sensors, if practical.
- the output of the sensor 34 is amplified in an amplifier 37 and produces the flow rate signal, E FLR , which may also be a 4 to 20 Ma current signal.
- the signals E Tin , E Tout and E FLR are applied to an input output (I/O) device 36, including analog to digital converters which digitize these signals, iteratively at a sampling rate, which may be at one second intervals.
- the digital signals are applied by the I/O 36 to a digital computer 38 which implements the control system which is schematically illustrated in FIG. 2.
- the computer communicates through the I/0 36 with the microwave generator and provides the control signal E c .
- This control signal is an analog signal, which as indicated above may vary between 0 and minus 1 volt.
- the microwave generator has suitable amplifiers which use this control signal E c to control a magnetron therein which produces the microwave energy.
- a digital to analog converter in the I/O 36 provides the analog control signal E c to the microwave generator 16.
- the I/O 36 includes memory (e.g., latches and other digital logic) for holding the digitized output signals, E Tout between sampling times.
- the computer 38 which implements the control system is preferably of the type suitable for real time multi-tasking operation (e.g., the Digital Equipment Company - PRO/380 or similar process control computer). This computer is programmed to carry out the processes or algorithms needed to perform feedback control, feedforward control, and adaptation processes which are shown in the flow charts (FIGS. 3A, B-1, B-2, C and D hereof).
- the term process is used herein with reference to the flowing fluid temperature control process.
- the term process also refers to computational processes as will be apparent from the context of the description as it proceeds.
- the microwave generator has a constant transfer function and can vary its output energy over a multikilowatt range. Absorption of the microwave energy by the bulk of the fluid is very rapid so that fast temperature transients can readily be followed and corrected by the energy supplied from the microwave generator.
- FIG. 2 illustrates the controller.
- This controller provides feedforward control in response to the temperature measured ahead (upstream) of the control point 12 by the sensor 28.
- the system provides feedback control in response to the temperature measured by the sensor 26 after (downstream) the control point.
- An adaptive processor 38 exercises control over a feedforward processor 40 in the feedforward loop.
- the adaptive processor 38 derives control signals from the feedback loop.
- the system is designed to control the temperature at the control point to a certain set point temperature T SP .
- the set point is adjustable by an input to the computer 38 from a keyboard. Operating constants of the control functions in the loops can be entered into the computer from the keyboard.
- the I/O 36, A/D may provide a 12 bit conversion for high accuracy and precise control.
- the feedback loop utilizes a feedback processor 42 which has an integral and proportional response which is illustrated as the summation operator 44 and the proportional operator 46 thereof (an integral function).
- the set point signal T SP is compared or summed as shown at 48 with the temperature signal E t out from sensor 26 to provide an error signal (e s2 ).
- This signal, e s2 is subjected to the proportional operator 46 and the integral operator 44 and summed at 50 to provide the feedback processor output x1.
- x1 can be positive or negative and respectively either adds to or subtracts from the output of the feedforward loop to control the amount of energy produced by the microwave generator 16, which is applied via the wave guide 22 to the microwave applicator 12 at the control point.
- the summation operator 44 is also affected by the adaptive processor 38 which provides a signal x4 to vary the integral value produced by the summation operator 44.
- the response of the feedback controller may be written as where T is the sampling interval (1 sec in this example), k denotes the k th sample and the other terms are defined above and K p , K I and K are tuning constants, discussed below.
- the feedforward processor has a proportional response in accordance with the control function of the process.
- the feedforward processor may also have a derivative response. This response is to the error signal e s1 obtained from the summation of the signal from the temperature sensor 28, E Tin and T sp at 41.
- the proportional response alone is used in this illustrative embodiment of the invention.
- This response may be expressed as follows: where a is the adaption gain of the control function, C p is the specific heat of the fluid in the pipe 10 in BTUs per pound per °F; W is the flow rate of the fluid in pounds per minute, ⁇ T is the temperature correction needed to bring the temperature at the control point 12 to the set point in degrees Farenheit; and K m is a constant which has a value of 57 when the above units are used.
- the derivative response if used, will be equal to where K d is the derivative response gain.
- the output of the feedforward controller x2 is a fractional value proportional to the desired output power from the microwave generator.
- the above equation for KW can be used to calculate the necessary output from the feedforward processor 40.
- the relationship between generator input and output may be nonlinear. Therefore, the best temperature control will be achieved when this nonlinearity is taken into account in determining X2. In other words, the system compensates for nonlinearity in the microwave generator.
- the output from the feedback controller 42, x1 may add to or subtract from the feedforward controller output x2.
- x1 and x2 are summed at 52 to provide a control output x c .
- This control output is converted from a digital signal into the analog signal E c in the digital analog converter D/A of the I/O 36.
- the analog output E c may vary from 0 to -1 volts (or other range depending upon the interface of the microwave generator) and controls the power generated by the microwave generator 16 so as to accurately and precisely regulate the temperature at the control point.
- the feedforward loop is responsible for changing the temperature at the control point by varying the amount of energy produced by the microwave generator 16.
- the system operates during transient changes and also adapts to changes in the process. For example, different batches of liquid may have different specific heats. Then the initial control function is incorrect. This control function is adapted or tuned by changing the adaptation gain a.
- the adaptation gain will also change to correct for other conditions such as a change in the frequency of the output of the microwave generator which may change the amount of energy transmitted to the fluid, which ultimately changes the controlled temperature.
- the microwave generator has negligible dynamics and responds substantially as a constant gain device.
- Changes in conditions of the process which affect the temperature variable are detected by the e s2 input from the feedback processor 42 to the adaptive processor 38.
- the criteria for adaptation to begin is when the rate of change of e s2 and the magnitude of e s2 reach a predetermined small value. This value may be a fraction, say 1/2 of the magnitude of the temperature control range of the system and 10% of such magnitude in terms of rate of change. For example, for a control range of ⁇ 1/2°F the signal e s2 representing the state of the system where process conditions are manifest, steady state condition, may be about 0.25°F in magnitude.
- the rate of change of e s2 may be approximately 0.025°F per second.
- x3 is derived in response to x2 by multiplying x2 by a value which depends upon the rate at which the control function is to be changed. In order to change the control function smoothly without hunting, it is desirable that the factor which x2 is multiplied by, be less than 1. A suitable value is 0.1. x3 then generates the new value of the tuning constant a in the control function which is used in the feedforward processor 40. Then changes in process conditions, such as a change in C p , which affects the tuning constant are manifest and the tuning constant can be adapted.
- any remaining error in the temperature (after the control point) is detected by the sensor 26 and further corrections are effected by the feedback processor 42.
- the combined output from both processors 40 & 42 makes up the input X c to the microwave generator 16.
- the output X2 from the feedforward processor 40 changes in response to adjustments in "a" (the tuning constant).
- the output X c remains constant so, for example, increases in X2 have to be accompanied by equal magnitude decreases in X1 etc.
- the output from the feedback controller X1 is due almost entirely to the integral response 44.
- the adaptive processor corrects the integral output from the integral operator 44 based upon the change in the feedforward output x2 by applying the signal x4 to the integral operator 44, thereby returning the loops to their balanced condition.
- X4 is given by the last term in the feedback controller response equation (1) where a suitable value for K is 0.1. The adaptation process continues on each iteration until the output from the feedback controller X1 is reduced to zero indicating near perfect feedforward controller tuning.
- FIG. 3A shows the overall program which consists of the feedback process (FB process) and the feedforward process (FF process).
- the program also has an adaptation process which principally generates the tuning constant x3.
- the program also has a process for calculating the output to the microwave generator 16 which produces x c .
- x c is a fractional value, proportional to the maximum output from the microwave generator, and is converted into an analog signal E c for controlling the generator 16 by the D/A 36.
- FIGS. 3B-1 and 3B-2 represent the feedback process.
- the program proceeds by determining whether a flag indicating that the system is set for manual rather than automatic operation. If the system is set for manual operation, the feedback output x1 is taken from the value which is set at an output location in memory (RAM) of the computer 38. As noted above, the end symbol means the end of the process, that is, that the output is ready.
- the determination is first made as to whether the set point has changed.
- Set point changes are entered into the computer into a location in memory from a keyboard.
- the feedback and feedforward processes detect a change in setpoint simultaneously. Under normal circumstances, the feedforward process supplies all the necessary control. Due to the exponential response of the RTD 26 and associated transmitter 30 (FIG. 1), the new value of process temperature is delayed in being transmitted to summing junction 48 (FIG. 2). Consequently, if the response of the feedback process was instantaneous, the temperature would then be overcorrected. Therefore, it is desirable to insert a delay in the response of the feedback process so that it does not respond immediately to a set point change. The delay is provided by a counter. Each sample time (e.g.
- the counter is decremented until the count is reduced to 0. Then the counter is reset to await the next delay cycle and the calculation of the feedback output proceeds. Of course, if there has been no set point change, the program jumps to the next calculation as shown on the flow chart.
- the next process is the calculation of the error signal e s2 .
- This signal is used in the adaptation process shown in FIG. 3D as shown by the off-page connectors A-A.
- the I/O 36 is programmed to sample the sensors and input new values into memory at each iteration (at the start of each scan).
- the calculation of the feedback output depends upon the tuning constants K p and K I . These constants are determined by the operator and stored in memory. They indicate the fractional part of the error signal e s2 used for the proportional and the integral responses, respectively.
- K p and K I are selected in accordance with the magnitude and rate of correction desired from the system. Similar adjustment technique for proportional and integral responses in control systems are described in classical papers in the field of control systems.
- the process calculates the feedback output x1 using the proportional value (from the proportional operator 46) and the integral value (from the integral operator 44).
- the feedback process proceeds via the connectors D-D to FIG. 3B-2.
- FIG. 3B-2 is a process which detects and compensates for what is known as "reset wind-up" by which is meant that the controller has reached 100% output and cannot make further corrections. In this system this means that the microwave generator has reached either 100% output (3 kilowatts) or is at 0% output. However, error signal e s2 may still exist for a period of time and could cause the integral value to increase to a large number which can only be reduced by having an error of opposite polarity existing generally for long periods of time. In this situation, the process is out of control. This mode of operation can be prevented as shown in FIG. 3B-2 by limiting the controller output and particularly the integral value to the value it has at the time the controller output reaches ⁇ 100%. Should the integral value attempt to exceed ⁇ 100%, it is set back to the corresponding maximum value. Accordingly, the system must use integral values at 0 or 100% of the controller output until the integral value decreases and control is returned to the system.
- the reset wind-up control program makes sure that the feedback output (FBO) and the integral value (I.V.) are set back either to +100% or -100% when the feedback output is out of limits.
- the program proceeds by checking the feedback output to determine whether it is in or out of limits. If the feedback output X1 is in limits, then it is used. If the feedback output is well out of limits, say 110%, an alarm may be sounded or other indicator of well out of range condition produced. If the feedback output is more than 100%, it is set back to 100%. Of course, if the feedback output indicates that it is not more than 100% but is out of limits, it is then set back to -100%.
- a similar procedure is used for setting the integral value back to 100% or -100%.
- the feedback output may be greater than 100% because of changes in the integral value alone. However, the out-of-limits condition can be due to the proportional value (P.V.).
- An integral value which may also be used may be less than the maximum value to accommodate for a contribution from the proportional value. If the integral value is in limits but the feedback output is out of limits, then the current integral value which is used in determining the feedback output is set to its value at the previous sampling (t s-1 ).
- FIG. 3C shows the feedforward process.
- the feedforward process calculates the output signal X2 according to the KW equation previously given and also includes a component in X2 which is proportional to the rate of change of incoming temperature.
- Signal X2 is also adjusted to compensate for any nonlinearity which may exist in the generator input-output characteristic as discussed above. When such compensation is used, it can be provided by approximating the E Tout f(E c ) curve by any suitable approximation technique. A simple example is a straight line segment approximation.
- the feedforward process is designed to calculate the x2 output in response to E FLR , the digital value of which is obtained from the I/O 36 (FIG. 1).
- the tuning constant "a" is also entered.
- the feedforward process calculates the derivative of the temperature measured ahead of the control point 12. This is the difference in temperature as measured at successive sampling times divided by the interval between sampling times (1 second in this embodiment).
- the e s1 signal may be used to calculate the temperature derivative.
- the e s1 signal is calculated from the set point and E Tin signal as shown at the summation point 41 in FIG. 2.
- the program then proceeds to calculate the kilowatts of microwave energy necessary to change the temperature to the set point temperature.
- This is a multiplication program which implements the formula for kilowatts given above.
- the kilowatt output is converted into a percentage of maximum power (3 kilowatts) to calculate the feedforward output x2. In other words, the kilowatt output is converted into a digital number representing a value from 0 to 100% where 100% corresponds to 3 kilowatts.
- FIG. 3D shows the adaptation process.
- the operator may, by keyboard control, render the adaptation process active or inactive. If the process is inactive, the program uses the values determined without adaptation (viz. x1, x2, and x c ). If the adaptation process is active, the absolute value of the magnitude of e s2 and the absolute value of the rate of change of e s2 , which is illustrated in FIG. 3D by a dot over the e s2 symbol, are checked to determine if they are within limits, d1 and d2.
- d1 and d2 are empirically derived. They may be selected in accordance with the criteria noted above and changed empirically when calibrating or trimming the control system on start-up of the process.
- adaptation or tuning constant value is calculated.
- the updated value of the tuning constant is calculated using the following equation: where a, x1, and x2 have been previously defined.
- the adaptation program also calculates the integral value adjustment x4 by multiplying x2 by a constant.
- the tuning value and the adjustment in the integral value are obtained only during steady state conditions, when changes in process conditions and manifested, rather than during dynamic or transient changes which are corrected by the operation of the feedforward and sometimes the feedback loop. Since the control system is adaptive, it maintains control over the process parameter of interest (temperature) notwithstanding changes in process conditions, such as a change in the specific heat of the fluid used in the process and without the need for perturbing the system.
Abstract
Description
- The present invention relates to process control systems, and more particularly to a control system (method and apparatus) which is adaptive on-line to changing process conditions.
- The invention is especially suitable for use in a process where the temperature of a flowing liquid requires regulation with a fast dynamic response to handle temperature transients while adapting to changing process conditions such as the specific heat of the flowing liquid, or other conditions affecting the temperature of the liquid. The process may be part of the coating of photographic paper or film which requires precise temperature regulation. The invention is applicable generally wherever a variable of a process is to be controlled, whether that variable be temperature or some other physical or chemical process variable; temperature control of flowing fluids as described herein being an illustrative and presently preferred embodiment of the invention.
- In regulating the temperature of flowing fluids, the conventional approach has been to use heat exchangers. Heat exchangers as are available are inadequate in providing the dynamic response for fast and precise regulation of the temperature of the fluid, for example, within a few seconds. In addition, heat exchangers generally are large in size, do not heat uniformly and must be maintained by cleaning and purging in order to keep them operative and avoid product contamination. Conventional control systems using heat exchangers for temperature control of flowing fluids have been inadequate for providing precise regulation while at the same time adapting (preferably in real time and on-line) to changing process conditions. Such conditions may, for example, be the specific heat of different batches of the fluids used in the process, such as the different batches of a chemical or different chemicals used in the manufacture of photographic products. Systems having heating means with faster dynamic response than heat exchangers and conventional control systems have been inadequate to provide the precise regulation which is needed by adapting to changing process conditions or compensating on-line and continuously for other changing factors which require a change in the dynamic response of the control system.
- Control systems sometimes have both feedforward and feedback control loops; feedforward control being responsive to cancel an incoming disturbance or change while feedback control being responsive to the result of the action which has been taken. Feedback and feedforward controllers have been combined, the feedback controller compensating for any error remaining after the feedforward action. See for example the following patents discussing feedback and feedforward controllers: Harrison, et al., U.S. 3,976,179, August 24, 1976; Koschmann, U.S. 4,290,986, September 22, 1981; Dibiano, U.S. 4,367,354, January 4, 1983; Peterman, et al., U.S. 4,543,245, September 24, 1985 and Dziubakowski, 4,776,301, October 11, 1988. It has also been suggested to provide adaptive control for tuning of the control law or control function of a controller. See the text "Adaptive Filtering - Prediction and Control" by G.C. Goodwin and K.S. Sin (1984), Prentice Hall - Englewood Cliffs, NJ, USA, especially pages 178-181, for adaptive control concepts.
- In known adaptive systems, adaptive control is implemented by perturbing the system (See Paz, U.S. 3,567,395 of March 2, 1971). Then the system response is detected and the system is adjusted or tuned. This approach is inadequate when the effect of these perturbations produces a variation beyond the limits of regulation required. For example, in a system requiring precise regulation of temperature of flowing fluids, particularly chemicals in a photographic film manufacturing process, the perturbation could change the temperature beyond the limits required by the process. Moreover, the approach is difficult to implement since both the amplitude and frequency content of the perturbing input must affect all response modes of the controlled process which are of interest. Otherwise, the control algorithm adaptation can diverge rather than converges. In other words, the system never reaches a state where the controlled variable is within the desired range, but rather is forced out of control; certainly an undesirable result.
- It is a feature of this invention to provide fast dynamic response which lends itself to adaptive control, in the case of a fluid flow temperature control system, using a microwave generator from which microwave energy is derived and applied to rapidly and uniformly heat and increase the temperature of the flowing fluid. Cooling with the microwave generator can be obtained by using an incoming fluid temperature lower than the desired (set point) temperature. Then, the microwave generator continually heats the fluid to desired temperature. Should the temperature of the incoming fluid increase, the output from the microwave generator is decreased, thereby causing less heat to be applied and providing a cooling action. The microwave generator is under feedforward and feedback control so as to precisely regulate the temperature.
- Another feature of the invention is to provide control of a variable using feedback and feedforward loops with adaptive control of the feedforward control function from the feedback loop. Specifically, in accordance with the feature, the feedforward controller (if conditions were ideal, which is not the case) would substantially eliminate any deviation of the controlled variable from the desired value. The feedback controller, acting as a backup, then reduces the remaining deviation, and the output from the feedback controller is a measure of the feedforward controller's inability to reduce the remaining deviation (called tuning error) when the system reaches equilibrium. In other words, the feedback output information has, in accordance with the invention been found sufficient to adapt or tune the feedforward controller to match changing process conditions, without having to perturb the controlled variable in order to derive the necessary adaptation or tuning signals. An additional advantage of feedback control of adaptation, in accordance with the invention, is that its implementation is far less complicated than with approaches requiring perturbation of the controlled variable. Since the controlled variable does not have to be perturbed, the system is tuned without driving the variable outside of the limits in which it must be maintained to meet process requirements.
- Accordingly, it is the principal object of the invention to provide a process control system which is fast acting and which can regulate a process variable within precise limits with an extremely fast response.
- It is a further object of the invention to provide a process control system for temperature regulation of flowing fluids utilizing microwave energy from a microwave generator so as to obtain extremely fast response (e.g. approximately 0.6 seconds to change the energy from zero to 100% or maximum output) and which does not affect the control function of the system by adding additional dynamics (the microwave generator acting essentially as a constant transfer function in the control law of the system).
- It is a further object of the present invention to provide an adaptive process control system using a feedforward controller; the control function or law of the sytem being adapted by a feedback controller, and the feedforward controller alone or together with the feedback controller providing an output for controlling the process variable of interest. The process variable may be temperature of a flowing fluid and a microwave generator which has an extremely fast response is responsive to the output of one or both of the controllers.
- It is a still further object of the invention to provide an adaptive process control system having a feedforward and a feedback controller with adaptive control of the feedforward controller in response to the output of the feedback controller, which controllers are implemented in a digital computer.
- Briefly described, the invention which may be embodied in a control system for temperature control by the application of microwave energy from a microwave generator which provides heat which changes the controlled variable. The system also includes feedforward and feedback controllers respectively responsive to the variable before and after a control point where the heat is applied from the microwave generator.
- Generally, the adaptive regulation of a process variable in accordance with the invention is carried out in a method having the following steps each of which may be performed with a digital computer. The steps are measuring the variable before and after a control point in the process; controlling the variable at the control point in response to changes in the variable as measured before the control point in accordance with a predetermined control function which relates the variable to the conditions of the process; and adapting the control function in response to changes in the variable as measured after the control point and when the changes represent changes in the condition of the process. The condition of the process being represented by changes, for example, when the variable is temperature, when the temperature error, as detected by the feedback controller, has such a low magnitude and slow rate of change as occurs when the process approaches stabilization.
- More specifically, the adaptive process control system has a feedforward controller. Suitably, this controller has proportional and derivative responses. In accordance with the preferred embodiment of the invention, only the proportional response is adapted to match changing process conditions. It will be understood that the derivative response or both proportional and derivative responses can be adapted. The system has a feedback controller having proportional and integral responses. The feedforward response may change in one sense, for example, to increase the temperature where the temperature is the controlled variable, while the feedback controller output can change in opposite senses. In other words, the feedforward conroller has a unipolar response whereas the feedback controller has a bipolar response which can add or subtract from that of the feedforward controller. Adaptation is obtained by changing the gain (a constant in the control function) of the feedforward controller in accordance with the feedback output. Preferably, the system operates iteratively, and is implemented with a digital computer having a sampling rate appropriate to accurately detect the fastest rate of change anticipated in the controlled variable. In the present embodiment, a sampling rate of one second may be used. The feedforward gain is adjusted to tune the feedforward controller at steady state conditions when the output rate and magnitude of the variable to which the feedback loop is responsive (the temperature after the control point) is well within limits within which regulation is desired. For example, in a system which regulates temperature within a ± 0.5°F, the limits may be 0.25°F in magnitude and 0.025°F per second in rate of change. Since tuning by adjusting the feedforward gain results in a change in the output from the feedforward controller, a compensating change is made in the output of the feedback controller when steady state conditions are detected. Thus, as the feedforward controller output provides a control signal which changes the controlled variable to its set point value, the output from the feedback controller approachs zero. Should process conditions change, for example a new batch of chemical having slightly different specific heat in an embodiment for temperature control of photographic chemicals, the feedforward tuning will be changed automatically by the feedback controller. No special perturbation of the system is required in order to obtain adaptive control.
- The foregoing and other objects, features, and advantages of the invention and the best mode now known for carrying out the invention will become more apparent from a reading of the following description in connection with the accompanying drawings in which:
- FIG. 1 is a block diagram schematically illustrating an adaptive process control system in accordance with the invention for control of the temperature of a flowing liquid;
- FIG. 2 is a functional block diagram illustrating in greater detail than FIG. 1 the system illustrated therein; and
- FIGS. 3A, 3B-1, 3B-2, 3C and 3D constitute a flow diagram illustrating the programming of the digital computer which implements the adaptive process control system shown in FIG. 2.
- Referring to FIG. 1 there is shown apparatus for carrying out a portion of a photographic film manufacturing process. The chemical being dispensed is a fluid (a liquid) which must be regulated precisely in temperature. The chemical is mixed in kettles (not shown) and passes through a system of plumbing including valves and other hardware which are also not shown. The portion of the process shown in FIG. 1 provides the necessary temperature control.
- The chemical flows through a
plastic pipe 10 past atemperature control point 12 in a waveguide applicator section 14, which applies microwave energy from amicrowave generator 16 to heat the chemical at the control point.Conductive couplings applicator 14 against the escape of microwave energy. Aplastic pipe 10 carrying the fluid extends through these couplings. - The
applicator 14 is tuned to the frequency of the microwave generator. For example, a suitable generator is made by Gerling Laboratories of Modesto, California, USA and produces an output frequency of 2.45 GHz. Other frequencies may be used, for example, when larger waveguides are practical. The microwaves are transmitted by awave guide 22 to theapplicator 14. Theapplicator 14 is tuned by a slug tuner 24. Such tuners are shown in U.S. Patent 4,689,459 issued August 25, 1987. The microwave generator is controlled by a control signal (for example, a voltage which may vary from 0 to minus 1 volt) which changes the microwave energy supplied to theapplicator 14 from 0 to 3 KW. - The temperature of the incoming fluid is measured ahead of the control point and also after the
control point 12 bytemperature sensors sensor 26 is substantially the same as the temperature of the fluid at the control point. The sensors are suitably resistance temperature devices (RTD) which are responsive to the temperature of the liquid in the pipe. The RTD's form one arm of bridge circuits. These bridge circuits are contained inRTD transmitters RTDs - A flow sensor 34 (FRS) is also disposed to sense the flow rate of the liquid through the
pipe 10. This flow sensor may be a Micromotion flow sensor. Such sensors have a vibrating loop with pickup coils. The loop vibrates at a frequency of vibration proportional to the rate of flow through the loop. Other sensors may be used such as vane sensors, if practical. The output of thesensor 34 is amplified in anamplifier 37 and produces the flow rate signal, EFLR, which may also be a 4 to 20 Ma current signal. - The signals ETin, ETout and EFLR are applied to an input output (I/O)
device 36, including analog to digital converters which digitize these signals, iteratively at a sampling rate, which may be at one second intervals. The digital signals are applied by the I/O 36 to adigital computer 38 which implements the control system which is schematically illustrated in FIG. 2. The computer communicates through the I/0 36 with the microwave generator and provides the control signal Ec. This control signal is an analog signal, which as indicated above may vary between 0 and minus 1 volt. The microwave generator has suitable amplifiers which use this control signal Ec to control a magnetron therein which produces the microwave energy. A digital to analog converter in the I/O 36 provides the analog control signal Ec to themicrowave generator 16. The I/O 36 includes memory (e.g., latches and other digital logic) for holding the digitized output signals, ETout between sampling times. Thecomputer 38 which implements the control system is preferably of the type suitable for real time multi-tasking operation (e.g., the Digital Equipment Company - PRO/380 or similar process control computer). This computer is programmed to carry out the processes or algorithms needed to perform feedback control, feedforward control, and adaptation processes which are shown in the flow charts (FIGS. 3A, B-1, B-2, C and D hereof). - The term process is used herein with reference to the flowing fluid temperature control process. The term process also refers to computational processes as will be apparent from the context of the description as it proceeds. The microwave generator has a constant transfer function and can vary its output energy over a multikilowatt range. Absorption of the microwave energy by the bulk of the fluid is very rapid so that fast temperature transients can readily be followed and corrected by the energy supplied from the microwave generator.
- FIG. 2 illustrates the controller. This controller provides feedforward control in response to the temperature measured ahead (upstream) of the
control point 12 by thesensor 28. The system provides feedback control in response to the temperature measured by thesensor 26 after (downstream) the control point. In other words, there is a feedforward controller or control loop and a feedback controller or control loop. Anadaptive processor 38 exercises control over afeedforward processor 40 in the feedforward loop. Theadaptive processor 38 derives control signals from the feedback loop. The system is designed to control the temperature at the control point to a certain set point temperature TSP. The set point is adjustable by an input to thecomputer 38 from a keyboard. Operating constants of the control functions in the loops can be entered into the computer from the keyboard. The I/O 36, A/D may provide a 12 bit conversion for high accuracy and precise control. - The feedback loop utilizes a feedback processor 42 which has an integral and proportional response which is illustrated as the summation operator 44 and the proportional operator 46 thereof (an integral function). The set point signal TSP is compared or summed as shown at 48 with the temperature signal Et out from
sensor 26 to provide an error signal (es2). This signal, es2, is subjected to the proportional operator 46 and the integral operator 44 and summed at 50 to provide the feedback processor output x₁. x₁ can be positive or negative and respectively either adds to or subtracts from the output of the feedforward loop to control the amount of energy produced by themicrowave generator 16, which is applied via thewave guide 22 to themicrowave applicator 12 at the control point. The summation operator 44 is also affected by theadaptive processor 38 which provides a signal x₄ to vary the integral value produced by the summation operator 44. The response of the feedback controller may be written as - The feedforward processor has a proportional response in accordance with the control function of the process. The feedforward processor may also have a derivative response. This response is to the error signal es1 obtained from the summation of the signal from the
temperature sensor 28, ETin and Tsp at 41. The proportional response alone is used in this illustrative embodiment of the invention. This response may be expressed as follows:pipe 10 in BTUs per pound per °F; W is the flow rate of the fluid in pounds per minute, ΔT is the temperature correction needed to bring the temperature at thecontrol point 12 to the set point in degrees Farenheit; and Km is a constant which has a value of 57 when the above units are used. -
- The output of the feedforward controller x₂ is a fractional value proportional to the desired output power from the microwave generator. The above equation for KW can be used to calculate the necessary output from the
feedforward processor 40. In practice, the relationship between generator input and output may be nonlinear. Therefore, the best temperature control will be achieved when this nonlinearity is taken into account in determining X₂. In other words, the system compensates for nonlinearity in the microwave generator. The output from the feedback controller 42, x₁, may add to or subtract from the feedforward controller output x₂. x₁ and x₂ are summed at 52 to provide a control output xc. This control output is converted from a digital signal into the analog signal Ec in the digital analog converter D/A of the I/O 36. Only the D/A of the I/O 36 is shown in FIG. 2. The analog output Ec may vary from 0 to -1 volts (or other range depending upon the interface of the microwave generator) and controls the power generated by themicrowave generator 16 so as to accurately and precisely regulate the temperature at the control point. The feedforward loop is responsible for changing the temperature at the control point by varying the amount of energy produced by themicrowave generator 16. The system operates during transient changes and also adapts to changes in the process. For example, different batches of liquid may have different specific heats. Then the initial control function is incorrect. This control function is adapted or tuned by changing the adaptation gain a. The adaptation gain will also change to correct for other conditions such as a change in the frequency of the output of the microwave generator which may change the amount of energy transmitted to the fluid, which ultimately changes the controlled temperature. The microwave generator has negligible dynamics and responds substantially as a constant gain device. - Changes in conditions of the process which affect the temperature variable are detected by the es2 input from the feedback processor 42 to the
adaptive processor 38. The criteria for adaptation to begin is when the rate of change of es2 and the magnitude of es2 reach a predetermined small value. This value may be a fraction, say 1/2 of the magnitude of the temperature control range of the system and 10% of such magnitude in terms of rate of change. For example, for a control range of ±1/2°F the signal es2 representing the state of the system where process conditions are manifest, steady state condition, may be about 0.25°F in magnitude. The rate of change of es2 may be approximately 0.025°F per second. When these values are detected in theadaptive processor 38, the process is sufficiently close to steady state and theadaptation processor 38 outputs its control signal x₃. x₃ is derived in response to x₂ by multiplying x₂ by a value which depends upon the rate at which the control function is to be changed. In order to change the control function smoothly without hunting, it is desirable that the factor which x₂ is multiplied by, be less than 1. A suitable value is 0.1. x₃ then generates the new value of the tuning constant a in the control function which is used in thefeedforward processor 40. Then changes in process conditions, such as a change in Cp, which affects the tuning constant are manifest and the tuning constant can be adapted. No perturbation of the system (no artificial changes in temperature) is necessary to adjust the tuning constant or to otherwise adapt the system for changes in process conditions. Accordingly, precise control which adapts for changes in process conditions or initial mistuning (or other faults) while exercising a fast dynamic response to transients in the parameter (the temperature) being controlled is maintained. - Any remaining error in the temperature (after the control point) is detected by the
sensor 26 and further corrections are effected by the feedback processor 42. The combined output from bothprocessors 40 & 42 makes up the input Xc to themicrowave generator 16. During the adaptation process the output X₂ from thefeedforward processor 40 changes in response to adjustments in "a" (the tuning constant). However, it is essential that the combined input Xc remain constant so, for example, increases in X₂ have to be accompanied by equal magnitude decreases in X₁ etc. At the near steady state condition where adaptation is active, the output from the feedback controller X₁ is due almost entirely to the integral response 44. Accordingly, the adaptive processor corrects the integral output from the integral operator 44 based upon the change in the feedforward output x₂ by applying the signal x₄ to the integral operator 44, thereby returning the loops to their balanced condition. X₄ is given by the last term in the feedback controller response equation (1) where a suitable value for K is 0.1. The adaptation process continues on each iteration until the output from the feedback controller X₁ is reduced to zero indicating near perfect feedforward controller tuning. - Referring to the flow charts and particularly to FIG. 3A. it will be seen that the flow charts have start and end points. The start points are the beginning of each scan or iteration. The end points means that an output value is available, for example, the feedback loop output x₁, the feedforward output x₂ or the adaptive processor outputs x₃ and x₄. FIG. 3A shows the overall program which consists of the feedback process (FB process) and the feedforward process (FF process). The program also has an adaptation process which principally generates the tuning constant x₃. The program also has a process for calculating the output to the
microwave generator 16 which produces xc. As discussed in connection with FIG. 2, xc is a fractional value, proportional to the maximum output from the microwave generator, and is converted into an analog signal Ec for controlling thegenerator 16 by the D/A 36. - FIGS. 3B-1 and 3B-2 represent the feedback process. The program proceeds by determining whether a flag indicating that the system is set for manual rather than automatic operation. If the system is set for manual operation, the feedback output x₁ is taken from the value which is set at an output location in memory (RAM) of the
computer 38. As noted above, the end symbol means the end of the process, that is, that the output is ready. - If the system is set for automatic operation, the determination is first made as to whether the set point has changed. Set point changes are entered into the computer into a location in memory from a keyboard. The feedback and feedforward processes detect a change in setpoint simultaneously. Under normal circumstances, the feedforward process supplies all the necessary control. Due to the exponential response of the
RTD 26 and associated transmitter 30 (FIG. 1), the new value of process temperature is delayed in being transmitted to summing junction 48 (FIG. 2). Consequently, if the response of the feedback process was instantaneous, the temperature would then be overcorrected. Therefore, it is desirable to insert a delay in the response of the feedback process so that it does not respond immediately to a set point change. The delay is provided by a counter. Each sample time (e.g. 1 second intervals) the counter is decremented until the count is reduced to 0. Then the counter is reset to await the next delay cycle and the calculation of the feedback output proceeds. Of course, if there has been no set point change, the program jumps to the next calculation as shown on the flow chart. - The next process is the calculation of the error signal es2. This signal is used in the adaptation process shown in FIG. 3D as shown by the off-page connectors A-A. It will be appreciated that the I/
O 36 is programmed to sample the sensors and input new values into memory at each iteration (at the start of each scan). The calculation of the feedback output depends upon the tuning constants Kp and KI. These constants are determined by the operator and stored in memory. They indicate the fractional part of the error signal es2 used for the proportional and the integral responses, respectively. The value of Kp and KI are selected in accordance with the magnitude and rate of correction desired from the system. Similar adjustment technique for proportional and integral responses in control systems are described in classical papers in the field of control systems. - The process calculates the feedback output x₁ using the proportional value (from the proportional operator 46) and the integral value (from the integral operator 44). However, the integral value (I.V.) at the preceding sample time is used ( at t = (k - 1)T) as indicated by the equation for X₁ (kT) given above. This is because the integral is the area under the es2 vs. time curve and a delay of 1 sample time is needed until the integral is obtained. Accordingly, the I.V. up to the previous sample time is used. Then for use in the next calculation the integral value at the current sample time (t = kT) is obtained. The feedback process proceeds via the connectors D-D to FIG. 3B-2.
- FIG. 3B-2 is a process which detects and compensates for what is known as "reset wind-up" by which is meant that the controller has reached 100% output and cannot make further corrections. In this system this means that the microwave generator has reached either 100% output (3 kilowatts) or is at 0% output. However, error signal es2 may still exist for a period of time and could cause the integral value to increase to a large number which can only be reduced by having an error of opposite polarity existing generally for long periods of time. In this situation, the process is out of control. This mode of operation can be prevented as shown in FIG. 3B-2 by limiting the controller output and particularly the integral value to the value it has at the time the controller output reaches ± 100%. Should the integral value attempt to exceed ± 100%, it is set back to the corresponding maximum value. Accordingly, the system must use integral values at 0 or 100% of the controller output until the integral value decreases and control is returned to the system.
- The reset wind-up control program, set forth in detail in FIG. 3B-2, makes sure that the feedback output (FBO) and the integral value (I.V.) are set back either to +100% or -100% when the feedback output is out of limits. The program proceeds by checking the feedback output to determine whether it is in or out of limits. If the feedback output X₁ is in limits, then it is used. If the feedback output is well out of limits, say 110%, an alarm may be sounded or other indicator of well out of range condition produced. If the feedback output is more than 100%, it is set back to 100%. Of course, if the feedback output indicates that it is not more than 100% but is out of limits, it is then set back to -100%. A similar procedure is used for setting the integral value back to 100% or -100%. The feedback output may be greater than 100% because of changes in the integral value alone. However, the out-of-limits condition can be due to the proportional value (P.V.). The maximum integral value is used (I.V. = 100%). An integral value which may also be used may be less than the maximum value to accommodate for a contribution from the proportional value. If the integral value is in limits but the feedback output is out of limits, then the current integral value which is used in determining the feedback output is set to its value at the previous sampling (ts-1).
- FIG. 3C shows the feedforward process. The feedforward process calculates the output signal X₂ according to the KW equation previously given and also includes a component in X₂ which is proportional to the rate of change of incoming temperature. Signal X₂ is also adjusted to compensate for any nonlinearity which may exist in the generator input-output characteristic as discussed above. When such compensation is used, it can be provided by approximating the ETout f(Ec) curve by any suitable approximation technique. A simple example is a straight line segment approximation. The feedforward process is designed to calculate the x₂ output in response to EFLR, the digital value of which is obtained from the I/O 36 (FIG. 1). The tuning constant "a" is also entered. This tuning constant comes from the adaptation process (see FIG. 3A). The feedforward process calculates the derivative of the temperature measured ahead of the
control point 12. This is the difference in temperature as measured at successive sampling times divided by the interval between sampling times (1 second in this embodiment). The es1 signal may be used to calculate the temperature derivative. The es1 signal is calculated from the set point and ETin signal as shown at thesummation point 41 in FIG. 2. The program then proceeds to calculate the kilowatts of microwave energy necessary to change the temperature to the set point temperature. This is a multiplication program which implements the formula for kilowatts given above. The kilowatt output is converted into a percentage of maximum power (3 kilowatts) to calculate the feedforward output x₂. In other words, the kilowatt output is converted into a digital number representing a value from 0 to 100% where 100% corresponds to 3 kilowatts. - FIG. 3D, shows the adaptation process. The operator may, by keyboard control, render the adaptation process active or inactive. If the process is inactive, the program uses the values determined without adaptation (viz. x₁, x₂, and xc). If the adaptation process is active, the absolute value of the magnitude of es2 and the absolute value of the rate of change of es2, which is illustrated in FIG. 3D by a dot over the es2 symbol, are checked to determine if they are within limits, d₁ and d₂. d₁ and d₂ are empirically derived. They may be selected in accordance with the criteria noted above and changed empirically when calibrating or trimming the control system on start-up of the process. If the system is not in steady state and there are large changes in es2 or the rate of change of es2, then adaptation is not used during this sample time. However, if the values of es2 and its rate of change are within limits, the adaptation or tuning constant value is calculated. The updated value of the tuning constant is calculated using the following equation:
- The adaptation program also calculates the integral value adjustment x₄ by multiplying x₂ by a constant. As noted above, the tuning value and the adjustment in the integral value are obtained only during steady state conditions, when changes in process conditions and manifested, rather than during dynamic or transient changes which are corrected by the operation of the feedforward and sometimes the feedback loop. Since the control system is adaptive, it maintains control over the process parameter of interest (temperature) notwithstanding changes in process conditions, such as a change in the specific heat of the fluid used in the process and without the need for perturbing the system.
- From the foregoing description, it will be apparent that there has been provided an improved adaptive process control system. The system has been described in connection with the control of a specific variable, namely temperature. It will be appreciated that the invention may be used to control other variables and the control device may then be of some other type than the herein illustrated microwave generator. For example, valves may be used as control devices, to add hotter or cooler liquid or different reagents to a process. Microwave energy has been found especially advantageous to provide rapid and precise temperature control of a flowing fluid. Other variations and modifications within the scope of the invention will undoubtedly suggest themselves to those skilled in the art. Accordingly, the foregoing description should be taken as illustrative and not in a limiting sense.
Claims (28)
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US408016 | 1989-09-15 | ||
US07/408,016 US5130920A (en) | 1989-09-15 | 1989-09-15 | Adaptive process control system, especially for control of temperature of flowing fluids |
Publications (3)
Publication Number | Publication Date |
---|---|
EP0417782A2 true EP0417782A2 (en) | 1991-03-20 |
EP0417782A3 EP0417782A3 (en) | 1992-10-14 |
EP0417782B1 EP0417782B1 (en) | 1996-08-14 |
Family
ID=23614503
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
EP90117651A Expired - Lifetime EP0417782B1 (en) | 1989-09-15 | 1990-09-13 | Adaptive process control system, especially for control of temperature of flowing fluids |
Country Status (4)
Country | Link |
---|---|
US (1) | US5130920A (en) |
EP (1) | EP0417782B1 (en) |
JP (1) | JPH03142501A (en) |
DE (1) | DE69028054T2 (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5302993A (en) * | 1992-01-22 | 1994-04-12 | Eastman Kodak Company | Microwave-heated film processing apparatus |
EP0677786A1 (en) * | 1994-04-15 | 1995-10-18 | Minnesota Mining And Manufacturing Company | Method for controlling the temperature of an intermittently standing and flowing fluid |
EP0802464A1 (en) * | 1996-04-18 | 1997-10-22 | Valeo Equipements Electriques Moteur | Method to regulate the exciting current of a automotive vehicle generator using digital processing and regulator device for carrying out this method |
CN100520660C (en) * | 2005-05-19 | 2009-07-29 | 南京大学 | Fast stabilizing temperature control method based on characteristic information |
US9582754B1 (en) | 2016-05-17 | 2017-02-28 | Roger Collins | Adaptive feed forward method for temperature control |
CN109932899A (en) * | 2017-12-15 | 2019-06-25 | 蓝星(北京)技术中心有限公司 | The optimal control method and system of reaction temperature in chemical products production |
Families Citing this family (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5209398A (en) * | 1992-09-02 | 1993-05-11 | Johnson Service Company | Model-based thermobalance with feedback |
US6192283B1 (en) | 1998-07-31 | 2001-02-20 | Siemens Energy & Automation, Inc. | Method and apparatus for adaptive control of a system or device |
JP2002538544A (en) * | 1999-03-03 | 2002-11-12 | エフエスアイ インターナショナル,インコーポレーテッド. | Method and apparatus for flow temperature control |
US6555994B1 (en) * | 2000-10-11 | 2003-04-29 | Eni Technology, Inc. | Predictive control of a generator output |
US6675073B2 (en) * | 2001-11-20 | 2004-01-06 | Steve Kieman | System and method for tuning the weight control of a flow of material |
US6658870B1 (en) * | 2003-01-07 | 2003-12-09 | Carrier Corporation | Absorption chiller control logic |
US7119312B2 (en) * | 2004-07-09 | 2006-10-10 | Sedlmayr Steven R | Microwave fluid heating and distillation method |
US7432482B2 (en) * | 2004-07-09 | 2008-10-07 | Sedlmayr Steven R | Distillation and distillate method by microwaves |
US20060049184A1 (en) * | 2004-08-23 | 2006-03-09 | Dti Innovations, Llc | Microwave-based hydronics heating system |
US20090134152A1 (en) * | 2005-10-27 | 2009-05-28 | Sedlmayr Steven R | Microwave nucleon-electron-bonding spin alignment and alteration of materials |
US7660660B2 (en) * | 2007-02-09 | 2010-02-09 | Toyota Motor Engineering & Manufacturing North America, Inc. | Systems and methods for regulation of engine variables |
US8812166B2 (en) * | 2009-10-20 | 2014-08-19 | Hewlett-Packard Development Company, L.P. | Supplying a resource to an entity from a resource actuator |
US9138982B2 (en) | 2011-04-27 | 2015-09-22 | Xerox Corporation | Image data based temperature control of a keyless inker |
US8508791B1 (en) | 2012-01-23 | 2013-08-13 | Xerox Corporation | Image feedforward laser power control for a multi-mirror based high power imager |
CN103605284B (en) * | 2013-11-14 | 2016-06-01 | 杭州电子科技大学 | The cracking waste plastics stove hearth pressure control method that dynamic matrix control is optimized |
CN105320174A (en) * | 2014-05-27 | 2016-02-10 | 齐鲁工业大学 | Vegetable waste fermentation temperature fuzzy control method |
US9871153B2 (en) | 2015-09-24 | 2018-01-16 | Elenion Technologies, Inc. | Photodetector with integrated temperature control element formed at least in part in a semiconductor layer |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US3791793A (en) * | 1972-01-31 | 1974-02-12 | Leeds & Northrup Co | Adaptive feed forward-feedback control of the concentration of a selected ion of a solution |
US4152567A (en) * | 1977-03-07 | 1979-05-01 | Mayfield Esther O | Microwave water heater |
GB2184526A (en) * | 1983-10-18 | 1987-06-24 | Gainsborough Electrical | Water heaters |
Family Cites Families (23)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US3601588A (en) * | 1966-05-23 | 1971-08-24 | Foxboro Co | Method and apparatus for adaptive control |
US3449215A (en) * | 1966-09-08 | 1969-06-10 | Phillips Petroleum Co | Distillation column control with biasing signal as feedback correction for computed product flow rate |
US3567395A (en) * | 1968-10-21 | 1971-03-02 | Phillips Petroleum Co | Apparatus for producing carbon black |
US3657524A (en) * | 1970-06-15 | 1972-04-18 | Ibm | Dual mode process control with continuous feedback during coarse mode |
US3778578A (en) * | 1971-11-10 | 1973-12-11 | R Long | Apparatus for producing super heated fluids |
US3758762A (en) * | 1972-07-10 | 1973-09-11 | Leeds & Northrup Co | Decoupled feedforward-feedback control system |
US3976179A (en) * | 1974-12-30 | 1976-08-24 | Texaco Inc. | Controlling the temperature of a depropanizer tower by chromatographic analysis of feed and bottoms |
US4006346A (en) * | 1975-11-13 | 1977-02-01 | Phillips Petroleum Company | Control method including feedforward signal |
US4290986A (en) * | 1978-08-03 | 1981-09-22 | Owens-Illinois, Inc. | Method for controlling a plastic extruder |
US4358652A (en) * | 1978-12-21 | 1982-11-09 | Kaarup Darrell R | Fluid heater apparatus |
US4257105A (en) * | 1979-05-02 | 1981-03-17 | Phillips Petroleum Company | Control of a cracking furnace |
US4310739A (en) * | 1980-05-19 | 1982-01-12 | Hatem John P | Fluid heater powered by microwave energy |
US4367354A (en) * | 1981-06-30 | 1983-01-04 | Phillips Petroleum Company | Temperature control of a selective hydrogenation process |
US4417116A (en) * | 1981-09-02 | 1983-11-22 | Black Jerimiah B | Microwave water heating method and apparatus |
US4543245A (en) * | 1981-12-30 | 1985-09-24 | Atlantic Richfield Company | Process for controlling temperature of a Claus sulfur unit thermal reactor |
GB2157459B (en) * | 1983-10-14 | 1987-02-11 | Ford Motor Co | Selective parametric self-calibrating control system |
AU560995B2 (en) * | 1984-02-07 | 1987-04-30 | Toshiba, Kabushiki Kaisha | Process control apparatus |
JPS6137261A (en) * | 1984-07-31 | 1986-02-22 | 菊地 真 | Heating apparatus for hyperthermia |
US4922412A (en) * | 1986-10-09 | 1990-05-01 | The Babcock & Wilcox Company | Apparatus and method using adaptive gain scheduling |
US4776301A (en) * | 1987-03-12 | 1988-10-11 | The Babcock & Wilcox Company | Advanced steam temperature control |
US4918584A (en) * | 1988-07-08 | 1990-04-17 | Performance Controls, Inc. | Self-adjusting servo device and method |
US4969084A (en) * | 1988-12-22 | 1990-11-06 | The Babcock & Wilcox Company | Superheater spray flow control for variable pressure operation |
US4978425A (en) * | 1989-06-13 | 1990-12-18 | Elsag International B.V. | Method for controlling the degree of cooking in a digester |
-
1989
- 1989-09-15 US US07/408,016 patent/US5130920A/en not_active Expired - Fee Related
-
1990
- 1990-09-13 EP EP90117651A patent/EP0417782B1/en not_active Expired - Lifetime
- 1990-09-13 DE DE69028054T patent/DE69028054T2/en not_active Expired - Fee Related
- 1990-09-14 JP JP2245997A patent/JPH03142501A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US3791793A (en) * | 1972-01-31 | 1974-02-12 | Leeds & Northrup Co | Adaptive feed forward-feedback control of the concentration of a selected ion of a solution |
US4152567A (en) * | 1977-03-07 | 1979-05-01 | Mayfield Esther O | Microwave water heater |
GB2184526A (en) * | 1983-10-18 | 1987-06-24 | Gainsborough Electrical | Water heaters |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5302993A (en) * | 1992-01-22 | 1994-04-12 | Eastman Kodak Company | Microwave-heated film processing apparatus |
EP0677786A1 (en) * | 1994-04-15 | 1995-10-18 | Minnesota Mining And Manufacturing Company | Method for controlling the temperature of an intermittently standing and flowing fluid |
US5605189A (en) * | 1994-04-15 | 1997-02-25 | Imation Corp. | Method for controlling the temperature of an intermittently standing and flowing fluid which is at rest during standstill intervals and is flowing during flow intervals |
EP0802464A1 (en) * | 1996-04-18 | 1997-10-22 | Valeo Equipements Electriques Moteur | Method to regulate the exciting current of a automotive vehicle generator using digital processing and regulator device for carrying out this method |
FR2747860A1 (en) * | 1996-04-18 | 1997-10-24 | Valeo Equip Electr Moteur | METHOD FOR CONTROLLING DIGITAL PROCESSING OF THE EXCITATION CURRENT OF A MOTOR VEHICLE ALTERNATOR AND REGULATOR DEVICE USING SUCH A METHOD |
US6081103A (en) * | 1996-04-18 | 2000-06-27 | Valeo Equipements Electriques Moteur | Method for regulating the excitation current in a motor vehicle alternator by digital processing, and a regulating device for carrying out the method |
CN1075683C (en) * | 1996-04-18 | 2001-11-28 | 瓦莱奥电机设备公司 | Arrangement and method for modulation of exciting current from ac generator of automobile |
CN100520660C (en) * | 2005-05-19 | 2009-07-29 | 南京大学 | Fast stabilizing temperature control method based on characteristic information |
US9582754B1 (en) | 2016-05-17 | 2017-02-28 | Roger Collins | Adaptive feed forward method for temperature control |
CN109932899A (en) * | 2017-12-15 | 2019-06-25 | 蓝星(北京)技术中心有限公司 | The optimal control method and system of reaction temperature in chemical products production |
Also Published As
Publication number | Publication date |
---|---|
DE69028054D1 (en) | 1996-09-19 |
JPH03142501A (en) | 1991-06-18 |
US5130920A (en) | 1992-07-14 |
DE69028054T2 (en) | 1997-01-23 |
EP0417782B1 (en) | 1996-08-14 |
EP0417782A3 (en) | 1992-10-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US5130920A (en) | Adaptive process control system, especially for control of temperature of flowing fluids | |
US4768143A (en) | Apparatus and method using adaptive gain scheduling algorithm | |
US4922412A (en) | Apparatus and method using adaptive gain scheduling | |
JP3949164B2 (en) | A feedback method for controlling nonlinear processes. | |
EP0710901B1 (en) | Multivariable nonlinear process controller | |
KR100684539B1 (en) | Method for controlling a gas flow within a digital mass flow controller, digital mass flow controller, method and system for generating a digitally enhanced flow rate signal | |
Yoshitani et al. | Model-based control of strip temperature for the heating furnace in continuous annealing | |
US4195337A (en) | Control method utilizing a model control scheme | |
US4814968A (en) | Self-tuning process controller | |
US5170341A (en) | Adaptive controller in a process control system and a method therefor | |
EP0710902A1 (en) | Method and apparatus for controlling multivariable nonlinear processes | |
US4358821A (en) | Method and apparatus for the incorporation of varying flow in the control of process quantities | |
EP0334698A2 (en) | Dead time compensated control loop | |
EP0329157B1 (en) | Film thickness controller | |
EP3816596A1 (en) | Regulation of rate of temperature change for a dry block calibrator | |
US4614860A (en) | Process and device for the anticipatory self-adapting regulation of a procedure | |
US20210251362A1 (en) | Method of controlling a haircare appliance | |
US5302993A (en) | Microwave-heated film processing apparatus | |
RU2243584C2 (en) | Method for optimal automatic adjustment of control system | |
JPH07261805A (en) | Automatic adjusting device for proportional plus integral plus derivative control parameter | |
RU2384872C2 (en) | Method for automatic optimal control system pulsed setup | |
KR970005554B1 (en) | Method of gain control using puzzy technique | |
JPH081330B2 (en) | Temperature controller for continuous fluid heating device | |
Venkatesan | A statistical approach to automatic process control (regulation schemes) | |
Tanneru | Experimental Evaluation and Characterisation of PID Controller For Spacecraft Thermal Testing |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PUAI | Public reference made under article 153(3) epc to a published international application that has entered the european phase |
Free format text: ORIGINAL CODE: 0009012 |
|
AK | Designated contracting states |
Kind code of ref document: A2 Designated state(s): DE FR GB |
|
PUAL | Search report despatched |
Free format text: ORIGINAL CODE: 0009013 |
|
AK | Designated contracting states |
Kind code of ref document: A3 Designated state(s): DE FR GB |
|
17P | Request for examination filed |
Effective date: 19930413 |
|
17Q | First examination report despatched |
Effective date: 19941027 |
|
GRAH | Despatch of communication of intention to grant a patent |
Free format text: ORIGINAL CODE: EPIDOS IGRA |
|
GRAA | (expected) grant |
Free format text: ORIGINAL CODE: 0009210 |
|
GRAH | Despatch of communication of intention to grant a patent |
Free format text: ORIGINAL CODE: EPIDOS IGRA |
|
AK | Designated contracting states |
Kind code of ref document: B1 Designated state(s): DE FR GB |
|
REF | Corresponds to: |
Ref document number: 69028054 Country of ref document: DE Date of ref document: 19960919 |
|
ET | Fr: translation filed | ||
PLBE | No opposition filed within time limit |
Free format text: ORIGINAL CODE: 0009261 |
|
STAA | Information on the status of an ep patent application or granted ep patent |
Free format text: STATUS: NO OPPOSITION FILED WITHIN TIME LIMIT |
|
26N | No opposition filed | ||
PGFP | Annual fee paid to national office [announced via postgrant information from national office to epo] |
Ref country code: GB Payment date: 19980806 Year of fee payment: 9 |
|
PGFP | Annual fee paid to national office [announced via postgrant information from national office to epo] |
Ref country code: FR Payment date: 19980902 Year of fee payment: 9 |
|
PGFP | Annual fee paid to national office [announced via postgrant information from national office to epo] |
Ref country code: DE Payment date: 19980928 Year of fee payment: 9 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: GB Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 19990913 |
|
GBPC | Gb: european patent ceased through non-payment of renewal fee |
Effective date: 19990913 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: FR Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20000531 |
|
PG25 | Lapsed in a contracting state [announced via postgrant information from national office to epo] |
Ref country code: DE Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES Effective date: 20000701 |
|
REG | Reference to a national code |
Ref country code: FR Ref legal event code: ST |